Skip to main content

Full text of "Program for better jobs and income: an analysis of costs and distributional effects : a study"

See other formats


y^ .ssy 



p 



,.% 



95th Congress 
2d Session 



JOINT COMMITTEE PEINT 



THE PROGRAM FOR BETTER JOBS AND 

INCOME: AN ANALYSIS OP COSTS AND 

DISTRIBUTIONAL EFFECTS 



A STUDY 

PREPARED FOR THE USE OF THE 

JOINT ECONOMIC COMMITTEE 
CONGRESS OF THE UNITED STATES 



.,,0"""^ 






-m 




FEBRUARY 3, 1978 



Printed for the use of the Joint Economic Committee 



21-890 



U.S. GOVERNMENT PRINTING OFFICE 
WASHINGTON : 1978 



For sale by the Superintendent of Documents, U.S. Government Printing Office 

Washington, D.C. 20402 

There is a minimum charge of $1.00 for each mail order 



JOINT ECONOMIC COMMITTEE 

n 30». 79th C 
RICHARD BOLLING, Missouri, Chairman 
I LOYD BENTSSN, Testa, Vkt Chairman 



HOUSE OF RBPRB8ENTATD 
BENE Dflin 

wili.iam B. MOO RBX AD, Pennsylvania 

II. HAMILTON. Indiana 
QILLI8 w. LONG, Louisiana 

O, PIKE, Mew Tort 
CLARENCE J. BROWN, Ohio 

.v BROWN, Michigan 
MARGARET M. HECBXER, Mseaeehosetti 
John H. ROU8SELOT, California 



SEN'ATF 
John BPARKMAN, Alabama 
WILLIAM PROXMIRE, W.^onsin 
ABRAHAM RIBICOPF, Connecticut 
BDWARD M. KENNEDY, Massachusetts 

Ith Dakota 
7ACOB K. JA\ its. New York 
william v. roth. Jr.. Delaware 
JAMES A. McCLURE, Idaho 
ORRIN O. hatch. Utah 



John r. Brass, Bsec 

Louis C. Kraithoff II, Airiftant Director 
Richard K. Kaufman, Ueneral Counsel 



O, Thomas Cator 

William A. I 

Thom \s f. DERNP.VRi. 



Economists 
Robk.RT I). H UfSXM 

kkm h. Hvossa 

L. DoUOLAfl Lee 



MlNORITT 

CsaSLSa EL Bradford vnr.s J. Fntin 

M. (' ITHSSDfl Miii.fr 



Philip McMirtin 

\h NoREi.i.i Man 
George r. Tyler 



roe D. Krumbbaar, Jr. 
Mark R. Policinski 



HI) 



LETTERS OF TRANSMITTAL 

January 31, 1978. 
To the Members of the Joint Economic Committee: 

Transmitted herewith for use of the members of the Joint Economic 
Committee and other Members of Congress is a study entitled "The 
Program for Better Jobs and Income: An Analysis of Costs and 
Distributional Effects." 

This is one of three studies commissioned by the Joint Economic 
Committee on the subject of welfare reform. These studies are in- 
tended to provide information and analysis to the Congress on this 
important issue. This study, prepared by Professors Robert Haveman 
and Eugene Smolensk}', University of Wisconsin, focuses on the budg- 
etary costs and distributional effects of varying certain basic elements 
in the Administration's welfare reform proposal. 

The views expressed in this study are those ot its authors and 
should not be interpreted as representing the views or recommenda- 
tions of the Joint Economic Committee or any of its members. 

Richard Bolling, 
Chairman, Joint Economic Committee. 



January 27, 1978. 
Hon. Richard Bolling, 
Chairman, Joint Economic Committee, 
U.S. Congress, Washington, B.C. 

Dear Mr. Chairman: Transmitted herewith is a study entitled 
"The Program for Better Jobs and Income: An Analysis of Costs and 
Distributional Effects," prepared by Professors Robert Haveman and 
Eugene Smolensk} r , University of Wisconsin. 

This study is the third Committee study on welfare reform intended 
to provide information and analysis on important aspects of the wel- 
fare reform proposal, including a review of its macroeconomic effects 
and an analysis of its budgetary costs and distributional effects. 

Drs. Haveman and Smolensky have evaluated the cost and benefit 
effects of various revisions of the Administration's proposals. 

The Committee is grateful for the cooperation and assistance of the 
U.S. Department of Health, Education, and Welfare in the preparation 
of this study. This study was reviewed by Deborah Xorelli Matz and 
Tom Cator of the Committee staff. 
Sincerety, 

John R. Stark, 

Executive Director, 
Joint Economic Committee. 

(in) 



Digitized by the Internet Archive 
in 2013 






http://archive.org/details/progbetterOOunit 



CONTENTS 



Page 
Letters of Transmittal __._ in 

THE PROGRAM FOR BETTER JOBS AND INCOME: AN 
ANALYSIS OF COSTS AND DISTRIBUTIONAL EFFECTS 

I. The budgetary costs of PBJI 1 

II. The effects of PBJI on various groups of people 3 

III. The cost and distributional effects of changing some PBJI charac- 
teristics 6 

IV. Summary and conclusion 12 

(V) 



THE PROGRAM FOR BETTER JOBS AND INCOME: AN 
ANALYSIS OF COSTS AND DISTRIBUTIONAL EFFECTS 

By Robert Haveman and Eugene Smolensky* 

The program for better jobs and income (PBJI) would change the 
pattern of income flows to a large number of the nation's families and 
would change both the incentives and the opportunities to work. In a 
previous report, we presented a critique of the entire plan. Here, we 
focus on two aspects of the proposal in more detail — its cost and its 
distributional consequences. 

In sections I and II, we briefly review what is now known about thft 
program's costs and distributional effects. This review is based largely 
on recent estimates by the Congressional Budget Office (CBO) and 
serves as background for some additional calculations made by the 
Department of Health, Education, and Welfare (DHEW), and sup- 
plied to us by the Joint Economic Committee. In these calculations, 
several aspects of the program were altered and the resulting changes 
in costs and distributional effects estimated. These particular changes 
were chosen because they appeared to be characteristics of the PBJI 
most likely to prove contentious during the legislative process. The 
results of these calculations are presented in section III. Finally, in 
section IV, we characterize what it is the administration is buying with 
the incremental expenditures required for PBJI, and summarize some 
of the findings from the simulations reported in section III. 

It should be emphasized that the data in this report were estimated 
by DHEW with the same basic microdata simulation model as was 
employed by the administration in their original description of the 
consequences of enacting PBJI. Our analysis is aimed at examining 
some of the effects of changing various aspects of PBJI; it does not 
challenge the accuracy or adequacy of the procedures by which DHEW 
predicts costs and benefits. 1 

I. The Budgetary Costs of PBJI 

The administration presented cost estimates at the time the details 
of the program were released. Table 1 presents the details of these 
estimates. The two main components of outlays are the cash benefits of 
$19.2 billion and the public service jobs of $8.8 billion. Offsetting these 
expenses are, primarily, the phaseout of three existing transfer pro- 
grams which accounts for $17.6 billion and the reduction which is 
possible in manpower training and other public employment programs 

♦The authors are professors of economics and staff members of the Institute for Research 
on Poverty, University of Wisconsin, Madison, Wis. 

1 The administration's cost and benefit estimates have been scrutinized in : Danziger, 
Sheldon ; Haveman, Robert ; and Smolensky, Eugene, "The Program for Better Jobs and 
Income — A Guide and a Critique," Joint Economic Committee Print, U.S. Congress, October 
17. 1977 ; Hausman, Leonard J. and Friedman, Barry L., "Work, Welfare, and the Program 
for Better Jobs and Income." Joint Economic Committee Print. U.S. Congress. October 11. 
1077 ; U.S. Congress, Congressional Budget Office, letter on cost estimates for Representa- 
tive James C Corman, November 29, 1977 ; and Storey, James R., et al., "The Better Jobs 
and Income Plan : A Guide to President Carter's Welfare Reform Proposal and Major 
Issues," the Urban Institute, mimeo, January 5, 1978. 

(1) 



186 of PBJI, which accounts for $6.9 billion. Considering both 
pluses and minuses, the net drain on the Federal budget in L9* 
mated to be $2.8 billion. 

Table l . . \ii PBJ /, a 

■■ :.!.! v I '». 1 7 



17.081 
nt* _ _- (15.31) 

participants plue adjustments . L( 
Administration 2. 20 



sts for in a trhing State supplements 1. f9 

tmenta for hold harmless, State share calculation, and Puerto Rico . 49 

■ 1 income tax credit l I. 50 

nee .61 

Employment program 8 v, » 



Full-time jobs 7 . ss 

Part-time jobs .52 

Administration .40 



tal outlays 31.08 

Savings from reductions in expenditure <>n other programs or incre 
in taxes 

Abolition of APDC 

Abolition of SSI 

Abolition of food stamps 

Reductions in EITC from additional earnings 

Reductions in CETA, Win, and l"I 

Reduction in housing programs 

Increased payroll taxes 

Reduction in fraud 

Wellhead tax 



«;. 


40 


5. 


70 


5. 


.')() 


1. 


10 


0. 


90 




30 




70 




40 


1. 


30 



Net cost of PBJI 2. 78 

IT*X benefits of $3,000,000,000 for those -who will not receive income supplements are 
n<>t considered by the administration to be a cost of the welfare program. 

The administration's cost estimate, in particular, the use of energy 
tax revenues and fraud elimination to onset program costs, and the 
neglect of additional tax reduction benefits given to middle income 
groups have been questioned. 2 However, they serve as a useful starting 
point for the analyses to he undertaken in part III. They were calcu- 
lated using the same computer model and are therefore consistent with 
and directly comparable to the numbers presented there. 

- Haveman, and Smolensky, ibid. ; U.S. Congress, Congressional Budget 
OftVo. ibid. ; and Storey et al., ibid. 



II. The Effects of PBJI on Various Groups of People 

One objective of the PBJI proposal is to integrate and improve the 
administration of and incentives created by the existing- melange of 
income transfer programs. A second objective is to increase the 
opportunities for, and necessity of, work for many who now are 
unemployed or underemployed. A third objective is to reduce the level 
of income poverty in the United States. For this reason, estimates of 
the effect of various program characteristics on groups of beneficiaries 
are relevant in the policy debate. 

Here we summarize some estimates of the distributional effects of 
PBJI, as produced by the Congressional Budget Office, employing a 
computer model very similar to that used by DHEW. These figures 
are to serve as a backdrop to our sensitivity-type analysis in part III 
in the same maimer as the base estimates of program costs presented 
in the previous section. 

Tables 2, 3, and 4 display the CBO estimates of the antipoverty 
and income distributional effects which PBJI would have achieved if it 
were in effect in 1975. The story which these figures tell can be sum- 
marized as follows : 

While two-thirds of welfare recipients had annual income below 
$5,000 in the current system, only 41 percent of assistance 
recipients would be below $5,000 under PBJI, a reduction of one- 
half million families. 

The current system eliminates $12.7 billion of the poverty gap, 
which is about 54 percent of the pre welfare gap. PBJI reduces 
the gap by $16.1 billion, or 68 percent. 

Under current welfare programs, 11.2 percent of all families are 
left in poverty, PBJI reduces this to 9 percent, a reduction of 
about 20 percent. 

PBJI appears to reduce poverty for most demographic groups — 
the aged, single parent families, intact families, families with 
disabled members, working poor families, and Southern families. 
It fails to raise the ratio of black to white incomes. 

YVhile about one-fourth of poor families would be made worse 
off under PBJI, 43 percent would experience an improvement in 
their financial status. Nearly all aged families would remain at 
least as well off. About 50 percent of single-parent families would 
be benefitted and very few left worse of}'. Over one-third of all 
black families are gainers, relative to about one-fifth of white 
families. 



21 v()0_7S- 



TABLE 2 DISTRIPU! -ITS BY PREWELFARE INCOME CUSSES UNDER CURRENT 

.ISTHATION'S WLLFARE REFORM PROPOSAL. 1975. 

|S $10,000 to Jib. to $25,000 and 

Pro,' J5.000 $9,999 S14.999 , Total 

16,310 18,327 8,548 74.576 

• e programs' 8,6'4 2.317 -1 168 12,715 

tamed income tax cred-L 3, 131 56 6, 483 

9.058 4~257 1 -7 934 2U 15.727 

ifare reform 
prot 
Cash assistance.. 9,382 3.934 2.-26 1,351 257 17.351 

PudIic service jobs 1.292 905 294 251 45 2.787 

Earned income tax - 4,783 4,741 1,432 134 13,129 

All components... 9. 5 7 57794 5.373 2,348 343 23.371 

'bilhons of dollars): 



Current pel 

•e Pro-ams' . 2.9 1.2 0.9 0.3 20.4 

ed income tax cre-M .5 .5 ? >0 » 1.2 

Total 15.5 3.4 1.2 .9 .3 21.6 

tration's welfare 
proposal: 
Cas- 17.1 4.7 2.1 1.2 .2 25.5 

Publx scvice jobs 3 3 1.8 0.6 . • .1 6.1 

Firned income tax credit . .4 IS 1.3 .4 >0 3.9 

All components 20.8 8.4 4.0 2.0 .3 35.6 

lei aid to families with dependent children, supplemental securty Income, state general assistance, and food 

-' Rounds tc ■ 

vnary estimated, Oct. \2, 1977, 

Source: Statement of Robert D. Re;schauer, "Preliminary Analyi.s of the Distributional Impact of the Administration's 
Aeltare Reform Proposal, " Task Force on Distributive Impacts of Budget and Economic Policy Committee of the Budget, 
U.S. Congress, Oct. 13. 1977. 



TABLE 3.— NUMBER OF FAMILIES IN POSTTAX, POSTTRANSFER POVERTY BY TYPE OF FAMILY AND REGION OF 
RESIDENCE UNDER CURRENT POLICY AND ADMINISTRATION REFORM PROPOSAL, 1975 

[Families in thousands] 



Posttax, posttransfer income 



Characteristics of families 



Post cash 

social 

insurance 




Administra- 
tion's reform 


income 


Current policy 


proposal 


10, 840 


8,339 


6,713 


2,916 
7,924 


2, 0-7 

6, 2G2 


1,444 
5,269 


2,577 
1,235 
1,058 
284 
1,676 
6,587 


1,565 

855 

541 

169 

1,213 

5,560 


1,172 
551 
454 
166 
523 

5, 017 


1,425 

9,415 


887 
7,^52 


721 

5,992 


2,305 

1,607 

912 

6,016 


1, 989 

1,200 

738 

4,412 


1,525 

1, 012 

587 

3,589 


8,039 
2,801 


5.2-8 
2,091 


4,854 
1,859 


4,250 
1,928 
2,207 
2, 454 


3,508 
1,307 
1,480 
1, 944 


2,935 
1,077 
1,064 
1,637 



All families 

Age of head: 

65 and over... 

Under 65 

Family type: 

Single parent with children 

Youngest child under 6 

Youngest child 6 to 13 

Youngest child 14 and over. 

2 parents with children 

Other 

Health status: 

Disabled member 

Nondisabled member 

Employment status of head: 

Working full time.. 

Working part time 

Unemployed 

Not in labor force 

Race of head: 

White 

Nonwhite 

Region of residence: 

South 

West 

Northeast 

North central 



Note: Preliminary estimates, Oct. 12, 1977, 
Source: See table 3. 



TABU 4._HUVbtR OF FAMILIES GAINING OR LOSING BENEFITS. BY FAMILY TYPE, UNDER ADVINISTRATION 
WELFARE REFORM PROPOSAL, 197S 

Amount of Income Amount of income 

lost gained 

TotH Families Total 

Current: v postal, posttransfer J50< r $100 to families with no famil es J 100 to $500 or 

income status $499 losing change gaining $499 more 

■ 

Below po^e: 1. 0*0 1.801 2.ril 2.902 1.303 1.599 

Abo.epo.- •■:. 1 i 2.835 50. PIC 14. 3(8 6.729 7,639 

Welfare status 

Cash assistance onl. 2C4 946 1.150 556 1.229 5.3 716 

Food stam, 1,492 i, 077 2.569 779 2.462 1.388 1074 

Cash assistance and food stamps 423 909 531 2.530 1,034 1,4% 

No cash assistance or food stamps . 2 55 52.621 11.048 5. C97 5,951 

Age of r-«ad 

m 741 1.228 10.915 3.312 2.008 1, 30* 

65 1.760 3.457 41.706 13.958 6.024 7.934 

III • : : child. en 522 1.070 2,321 3.691 1,492 2.198 

Youngest child under 6 157 141 293 385 1.59-i 884 

Youn. 241 507 I : " 1.652 614 1,038 

139 829 168 276 

2 parents .... 320 873 16.608 9.029 4,000 5,028 

1,033 1.659 2.742 33.691 4.550 2.540 2.010 

led member. . . . 651 453 1,114 1,123 1.261 540 721 

'.ondisabled member . 1.533 2.039 3,571 51,498 16.009 7.492 8,517 

• status of head: 

■ if full t me 519 683 1,202 30. 15 r , 8. 165 3.61.5 4,514 

,g part t rr,e 361 433 79: 7.358 2.235 1.05'. 1,181 

Unemployed 99 189 1.5*5 1.071 387 684 

labor force. 1,205 1,197 2 I 5.798 2.976 2,858 

Race of K ead: 

1,629 1,977 3.605 47.787 13.981 6.611 7,340 

NonA- te .. 555 524 1.079 4,334 3.289 1.391 1,897 
Region of residence: 

Sou! 531 1,017 1.518 15.586 6.388 3.318 3, 04Q 

West 591 401 991 10.260 2.677 1.139 1,483 

IhttSl 608 1,035 11.989 3.899 1,618 2.231 

North central... 1,110 14.636 4,306 1,877 2,429 

Allfarrles 2,181 2,501 4.685 52,621 17,269 8,032 9,237 

i j'vey of Income and Education underestimates the amount of food stamp benefits provided in 1975. Therefore, 
these r-el Tiinary estimates may overstate the number of gamers and understate the number of losers for those who 
'3np benefits under th-> current program. 

• rainitto estmates.Oct. 12. 1977 

Source See table 3. 

III. The Cost and Distributional Effects of Changing Some 
PBJI Characteristics 

3 ' ons I and IT serve as background For a supplemental Bet of 
calculations requested by tho Joint Economic ( Jommittee and supplied 
by DHEW. The purpose of these calculations is to analyze tin 1 effect 
on both program costs and distributional impact- of changes in a 
limited number of key program parameters. The analysis fomsses on 
those parameters likely to la* questioned during legislative delibera- 
t ons on the proposal. In tln> section, we describe the parameter 
changes and summarize their cost and distributive impacts. 

l. Elimination ol Two Tiers in the Cash Benefit Portion of the Fair ml 
gram and Movement to a Single Tier Negative Income Tas With a 

- change is equivalent to raising the guarantee and eliminating 
the $3,800 disregard when calculating cash supplements for those 
expected to work. It is a simplification of the plan. Some judge that 
this change will make the tn-k of getting people to -< i «'k jobs more 

ill by redu< of the penalties tor not work 

fers and jobs could easily becom - two quite separate programs if bb - 
\ ere made. 



$19.2 


$20 


1.175 


1.116 


$8.8 


$4.8 


$4.5 


0) 


3.8 


(') 


1 


(') 



The effects of this change in the structure of PBJI are shown in 
table 5. The implications of this change are very modest — while cash 
benefits rise by 4 percent, the cost of the jobs program falls by 5 
percent, leaving a net increase in program costs of $0.4 billion. If the 
impact of the two-tier provision on employment and work effort is as 
insubstantial as these simulations show, the program simplification 
achieved by this change should be seriously considered. In addition, 
estimates of the effect of the change indicate that the number of 
existing welfare recipients made worse off will not increase. Indeed, 
given the nature of the change, the increase in costs is likely to yield 
some increased poverty reduction as well as simplifying the proposal. 

TABLE 5.— THE EFFECT OF PARAMETER CHANGE 1 ON SELECTED COST AND DISTRIBUTIONAL INDICATORS 

Proposed Modified 

PBJI PBJI 

Federal cash benefits (billions) - — 

Number of job slots (millions) 

Cost of jobs program (billions) 

Cost of EITC (billions) 

Number of current A FDC recipients made worse off (millions) 

Number of current SSI recipients made worse off (millions) 

J Indicates no change from the proposed PBJI. 

2. Retention of the Existing Earned Income Tax Credit (EITC) or 
Completely Eliminating It 

The current EITC simultaneously reduces the benefit reduction 
rate for low earnings (largely, part-time) workers and adds 10 points 
to the benefit reduction rate for workers who earn from $4,000- 
$S,000 per year. As a result, work incentives are increased for the for- 
mer group, and decreased for the latter group. And, because of the 
shape of the distribution of earnings, the latter group is relatively 
larger than the former, very low earnings group. Two alternatives 
are available for reducing the disincentives problem for the higher 
income group. They are: (1) Increase the kink-point to the break- 
even income level, or (2) eliminate the EITC altogether. The choice 
made by the administration (to shift out the kink-point) reduces 
the share of total PBJI benefits going to the poor although their 
total benefits are increased. Table 6 indicates what is gained and at 
what cost. If the EITC is eliminated, the incentive for individuals 
to seek private rather than special public service employment is lost. 
The implication of this change on the demand for public service jobs 
and the characteristics of those who would hold them is shown by 
the simulation. 

TABLE 6.— THE EFFECT OF PARAMETER CHANGE 2 ON SELECTED COST AND DISTRIBUTIONAL INDICATORS 

PBJI with 
Proposed existing PBJI with. 

PBJI EITC no EITC 

Federal cash benefits (billions) 

Number of job slots (millions) 

Cost of jobs program (billions) 

Cost of EITC (billions) 

Number of current AFDC recipients made worse off (millions) 

Number of current SSI recipients made worse off (million)... 

1 Indicates no change from the proposed PBJI. 



$19.2 


0) 


(>) 


1.175 


1.199 


1.257 


$8.8 


$8.98 


$9.4 


$4.5 


$0.5 





3.8 


4.1 


4.4 


1 


(') 


(') 



Table B shows the impart of both changes in PBJ] an some Im- 
portant cos( and distribution variables. If expansion of the EIT(' 

were rejected in favor of maintaining the existing tax credit, the de- 
mand for public jobs would bo expanded as individuals would find 
private sector employment less attractive. However, this expansion 
i- small job- — implying a 2-percent increase in the costs of 

the public jobs program. The costs of tin- EITC would fall by about 
90 percent, a reduction of $4 billion in the budgetary costs of the pro- 
gram. Overall, a budget cost 3aving of $3.8 billion would be experi- 
enced. The effects of this budget cut are: to increase the demand for 
title IX special public service jobs, and some increase in the benefit 
reduction rate at earnings levels between $4,000 and $lf>,000. 

Complete elimination of the EITC 1 would have similar, but larger, 

effects. The demand for special public service jobs would increase by 

82,000 and the budgetary COSt of the jobs component would increase 
>.6 billion, or 7 percent. The budgetary cost of the total program 
would decrease by $3.9 billion, approximately the same amount as 
simply retaining the existing EITC. Moving from retention of the 
existing EITC to it-- elimination appears to yield very limited gains: 
trivial budget savings are experienced, an additional 60,000 jobs must 
be provided, and any gain in work incentives in the $4,000-$8.000 
range are offset by reduction- in the income range below $4,000. 
Neither of the changes analyzed have much effect on the status of the 
existing welfare population although in both cases the number of 
current AFDC recipients made worse off increases somewhat. By the 
nature of the changes, the target efficiency of the program would be 
increased as the primary reduction in costs is from reduced benefits 
accruing to nonpoverty families. Retaining the existing EITC results 
in greater poverty reduction than eliminating the EITC 1 altogether. 
The expanded EITC reduces poverty even more than does the exist- 
in- EITC. 



■- 



3. Increasing the Incentive To Take a Regular Public or Private 
Sector Rather Than a Special (Title IX) Public Sector Job 

In the original PBJ I, incentives to seek private sector employment 
rather than a public job came from two sources — the EITC paid on 
only private earnings and a lower cash benefit schedule for those on 
special purpose job.-. 3 When the program was finally unveiled, the full 
burden of inducing private 4 sector job search fell entirely on the EIT( \ 
The effect of this reduction in inducement for regular employment is 
shown in 'Fable 7. Substitution of the earlier, larger private sector in- 
ducement would reduce the costs of both the cash benefits and the 
jobs components of PBJ I. Taken together, a cost saving of $1 .6 billion 
would be experienced. Moreover, the number of new title IX jobs 
which would have to be created by the administration would be re- 
duced by IT):;, 000. as this number of worker- would choose private or 
regular public sector jobs. The primary gain from the $1.6 billion 
increa ociated with this change is a reduction in the complexity 

of calculating cash i> nefits. Because tin 4 small reduction in cash bene- 
fits (from $19.2 to Sis. 7 billion) reflects reduced payments to workers 
taking regular employment, the change will have little, if any, effect 
on the status of i risting welfare recipients. 

a In tiio earlier benefit Bchedule the fard od pul 



Proposed 


Modified 


PBJI 


PBJI 


$19.2 


$18.7 


1.175 


1.022 


$8.8 


$7.7 


$4.5 


$4.6 


3.8 


W 


1 


« 



TABLE 7.— THE EFFECT OF PARAMETER CHANGE 3 ON SELECTED COST AND DISTRIBUTIONAL INDICATORS 



Federal Cash Benefits (billions) 

Number of Job Slots (millions) 

Cost of Jobs Program (billions) 

Cost of EITC (billions) 

Number of Current AFDC recipients made worse off (millions) 

Number of current SSI recipients made worse off (million) 

1 Indicates no change from the proposed PBJI. 

4. Increasing the Wage Rate in the Special Public Service {Title IX) 
Jobs From the Minimum Wage to $3.00 ($4.00) Per Hour 

As the wage rate on the public service jobs rises, demand for them 
increases, particularly among those individuals now currently em- 
ployed full time, full year in the private sector. If the $8.8 billion cap 
on expenditures is maintained, total expenditures for cash assistance 
would probably rise. There are two partially offsetting effects at work. 
On the one hand, many more public service jobholders would receive 
no cash assistance at all. On the other hand, many more families would 
receive the maximum guarantee of the upper tier, if the expenditure 
cap limits the number of jobs available. 

Currently, there is mounting pressure for an increase in the wage 
rate paid for title IX jobs. The concern is that payment of the mini- 
mum wage would tag these jobs as second class, and perhaps more 
importantly, payment of the minimum wage would tend to under cut 
the prevailing wage rate in some labor markets. While these points 
have merit, increasing the wage payment has the potential for greatly 
increasing the demand for special public jobs and increasing the total 
cost of the program. Table 9 presents the implications of increasing 
the wage rate to $3 and to $4 per hour, under the assumption that the 
total demand for special public jobs will be met. 

The T3sults in table 8 are most revealing. The modest wage rate 
increase from $2.65 per hour to $3 per hour increases the demand for 
title IX jobs by 340,000, and increases the costs of the jobs component 
of PBJI by $3.5 billion. The increase in the total cost of PBJI is a 
smaller $2.6 billion because of reductions in cash benefits and the 
EITC. As the wage rate is moved up to $4 per hour, the changes in 
costs become much larger. The number of workers now preferring a 
special title IX job is increased from 1.175 million to 2.491 million — 
an increase of 1.316 million jobs. Without a cap the budget cost of the 
jobs component more than triples — from $8.8 billion to $26.9 billion — 
and the total cost of the program increases by $15.3 billion. The gains 
from an increase in the wage rate are real. They include avoidance of a 
stigma placed on the public jobs and the potential erosion of prevailing 
wage rates. However, the budgetary costs of increasing the wage rate 
to the $3 level and beyond are substantial. Although this change 
primarily affects the balance of workers between private and public 
sectors, there is some improvement in the economic status of existing 
AFDC recipients. 



LO 

TABLE 8. -THE EFFECT OF PARAMETER CHANGE 4 ON SELECTED COST AND DISTRI 3U T 1 0', AL 

INDICATORS 

Propped Modified Modified 
PBJI P8JI ($3.00) PBJl (J4.00) 

Federal cash benefits (billions) .... $19.2 $18 5 $17 4 

•r of job slots (millions) 1.175 1.516 2.491 

Cost of jobs pneraTi (billions) $8.8 $12.3 $26.9 

Costof EITC (I $4 5 $1 3 $3 5 

Number of current AFOC recipients made worse off (millions). 3.8 3.4 3 

Number of current SSI recipients made worse off (million). 1 (i) (i) 

i Indicates no change from the proposed PBJI. 

Mdki a Title IX Public Service Job Availabb to ''■> Primary 

/TV? . , i \ ! I 1 1 . . . i . . /. . .1 . 1 J ' »> i*<« 



• '. 



Earner in All Household I' nits 



PBJ] guarantees a public service job to the primary earner in nil 

household units with children: couples and unrelated individuals arc 
excluded from participating in the jobs component of the program. 
It would seem t<> he only B matter of time before these households are 
brought more fully into the system. Because unrelated individuals 
are concentrated at the low end of the earnings distribution, bringing 
them fully into the system could greatly increase the demand for public 
service jogs. One factor moderating the demand for jogs is that a large 
part of this population i- disabled and/or institutionalized. 

Table 9 illustrates the effect of this parameter change. As expected, 
the increase in the demand for public service jobs is substantial — an 
increase of 460 percent. The budgetary costs of the jobs component 
rise from $8.8 billion to over $45 billion. The total co>t of PBJ] with 
tin- modification is $37 billion greater than the administration's 
proposal. While expanding the coverage of the jobs program to include 
unrelated individuals and childless couples would increase the hori- 
zontal equity of the program, it entails large increases in budgetary 
costs. 

TABLE S.-THE EFFECT OF PARAMETER CHANGE 5 ON SELECTED COST AND DISTRIBUTIONAL INDICATORS 



Federal cash benefits (billions) ........ 

Number of job slots (millions) ._. 

Cost of jobs program (Dillions) 

Costof EITC (b'lhons). 

Number of current AFDC recipients made worse off (millions) 

N umber of current SSI recipients made worse off (million) 

1 Indicates no change from the proposed PBJI. 

6. Capping thi Jobs Component of PBJI by Imposing a Ceiling oj 

800,000 New Jobs 

The creation of new jobs on a ma— basis is a difficult undertaking. 
As we stated in our earlier study: 

mass creation <»f public service jobs for low wage-low skill workers is some- 
thing with which this country h:is no pn vious experience. The effort is analogous 
to :i private firm's promise to introduce a oew product, the manufacture of which 
requires a technology which has not yel been developed. In all ^M'• , ' cases, the 



Proposed 


Modified 


PBJI 


PBJI 


519 2 


$18.3 


1.175 


6.580 


$8.8 


$47.1 


54 5 


$4 5 


3.8 


V) 


1 


.9 



11 

effort is fraught with uncertainty, and the possibility of an ineffective and un- 
productive program must not be neglected. 4 

Given the difficulties of locating qualified contractor-sponsors and 
arranging productive work arrangements, it would not be surprising 
if the full complement of 1.4 million jobs could not be created during 
the first few years of the program. The number of jobs could also be 
constrained below the administration proposal, since for budgetary 
reasons, a lid may be placed on the number of jobs to be funded. 

Because of the structure of PBJI, limiting the number of jobs made 
available will not result in proportional reduction of total program 
costs. While some of the workers who would have received a public job 
will find alternative private sector employment, some will remain 
unemployed and fall back on the benefits from the cash transfer 
component of the program. 

Table 10 presents the cost consequences of limiting the number of 
job slots to 800,000. As expected, the budgetary costs of the jobs 
component is reduced — from $8.8 to $6.2 billion, a savings of $2.8 
billion. However, some of this saving is offset by a $0.5 billion in- 
crease in cash benefits and a $0.1 billion increase in theEITC. The net 
saving is S2 billion. Accompanying this saving, however, are the un- 
desired side effects of increasing the discretion of program adminis- 
trators, "cream-skimming" in the selection of applicants, and hori- 
zontal inequities in the allocation of jobs. 

TABLE 10.— THE EFFECT OF PARAMETER CHANGE 6 ON SELECTED COST AND DISTRIBUTIONAL INDICATORS 





Proposed 
PBJI 


Modified 

PBJI 


Federal cash benefits (billions)... 


$19.2 


$19.7 


Number of job slots (millions) 

Cost of jobs program (billions) 

Cost of EITC (billions).. 


1.175 

$8.8 

$4.5 


.8 
$6.2 
$4.6 


Numbercf current A FDC recipients made worse off (millions) 

Number of cu;rent SSI recipients made worse off (million) 


3.8 

1 


4.1 
(') 









1 Indicates no change from the proposed PBJI. 

7. Eliminate Federal Sharing of State Supplementation Costs and Use 
the Budgetary Savings To Increase the Guarantee on Both Tiers of PBJI 

Federal incentives to encourage State supplementation of benefits 
significantly complicate the structure of PBJI, compromising the 
administration's claims that the system has been made simpler and 
that horizontal equity has been increased. The accompanying restric- 
tions on the States also strain traditional intergovernmental relations. 
To the extent that the objective is to relieve the fiscal burden on the 
States, fiscal relief could also be obtained if the financial incentives 
for State supplementation were dropped and the Federal funds were 
put into higher guarantee levels. Were that done, and if States 
failed to supplement benefits on their own, many current recipients 
would be made worse off and the fiscal relief provided would be 
concentrated in the States now providing relatively low benefits. 
The effects on several relevant variables of simplifying the plan in 
this way are illustrated in table 11. 

* Sheldon Danziger, Robert Haveman, and Eugene Smolensky, op. clt. 



$19 2 


RO. 1 


1 175 


1 21 


$8 8 


59 1 


** 5 


J5 


3.8 


5.4 


1 


1.24 



12 

TABLE 11.— THE EFFECT OF PARAMETER CHANGE 7 ON SELECTED COST AND DISTRIBUTIONAL INDICATORS 

Proposed 

PBJI PBJI 

Federal cash benefit! (billions) 

Numbc- of :ob slots (millions) . 

Cost of jobs program (billions) 

Cost of EITC I 

Number of current AFDC recipients made worse off (millions). 

Number of current SSI recipients made worse off (millions)... . . 



In this simulation, costs of the Federal cash benefits portion of the 
bill were allowed to rise by SO. 8 billion. Nevertheless, both the 
of tli" jobs program and the costs of the EIT< ' also rise by a total of 
$0.8 billion. Despite the greater cost, the Dumber of current AFDC 
recipients made worse off increases by 42 percent , and the number of 
SSI recipients made worse off increases by *24 percent. Hence, simpli- 
fying the State supplements portion of the bill in this way has signifi- 
cant adverse distributional effects. And while some fiscal relief is 
provided to State governments by this modification, it- level i- sub- 
stantially reduced from that in the PBJI. 

IV. Summary and Conclusion 

The program for better jobs and income directly addresses the 
judgment of many observers of the current welfare system that those 
who cannot meet their basic Deeds through earnings should have cash 
assistance, but that those who can be provided the incentive and the 
opportunity to earn their way. Drafting a program to accomplish this 
objective is technically difficult. Providing cash assistance creates an 
incentive to some to reduce their work effort, and it is difficult to con- 
fine this work disincentive only to those who are judged unable to 
generate sufficient income through work. There are only two viable 
alternatives for minimizing the disincentive effects of cash transfers. 
One i< to enforce a work test through tough administration. This re- 
quires a large bureaucracy and considerable intrusion into the privacy 
of cash assistance recipients. The other alternative is to create effective 
opportunities for and financial incentives to work, or at least to reduce 
the substantia] disincentives present in existing program-. In the main. 
PB.II opts for this alternative. When work is refused by those expected 
to work, not only are earnings sacrificed, but the family sacrifices 
$1,900 per year in cash assistance. However, reliance is not entirely on 
opportunities and incentives, since the decision of who is and who is 
not expected to work is made by program administrators. 

PB.II could have relied on the $155 per month penalty to send those 
who can work into the private job market. However, recognizing the 
hardships that might thereby be created, a special public service jobs 
program is created. To keep the costs of the jobs program down, n 

based on the minin am wage laws, not at prevailing market levels. 
Because the market wage of a Large number of PB.II beneficiaries is 
this minimum wage, the special public service jobs would be attractive. 
However, these special jobs are intended to temporarily supplement 
private sector jobs. They are not intended to substitute for private 
jobs. PBJI participants could be moved out of the public and 
into the private sector by administrative procedures, but consistent 



13 

with the general approach, financial incentives are brought to bear. 
Through the EITC, work in the private sector or in a regular public 
sector job is given a financial bonus. 

Providing cash assistance, financial work incentives, work opportu- 
nities, financial incentives to seek private sector employment, and 
maintaining budgetary restraint make for a complicated program. Jn 
addition, PBJI seeks to grant fiscal relief to the States and to sustain 
current benefit levels for the vast majority of current welfare recipients 
in a way that will not jeopardize other objectives, and thus, the pro- 
gram becomes even more complicated. The technical problem of bal- 
ancing all these objectives involves a multiplicity of tradeoffs. How 
the framers of PBJI chose to trade off among these objectives has been 
illustrated in the preceding tables. 

Holding the wage on special public service jobs to the minimum 
wage is clearly an effective check on the demand for those jobs and 
on total program costs. (See table 8.) It does, however, constrain the 
attractiveness of the work opportunities provided. Restricting public 
service jobs to families with children also substantially reduces the 
demand for those jobs and hence controls program costs. (See table 9.) 
Again, however, there is a cost. The work opportunities provided are 
limited to one group in the population, creating some horizontal 
inequities. The program is designed to grant all eligible persons a job. 
This decision entails higher budgetary costs than a more restricted 
and less equitable (though perhaps more realistic) jobs program. (See 
table 10.) 

The complications associated with the two-tier cash assistance 
benefit schedule also stem from concern with work incentives. This 
concern may be exaggerated, as the elimination of the two- tier struc- 
ture increases cash benefits by only 5 percent and reduces the demand 
for special purpose jobs by 5 percent. (See table 5.) The expanded 
EITC in the program is also designed to increase the reward to 
work — in this case work in regular employment. The budgetary costs 
of PBJI are increased substantially by this provision, which modestly 
affects the demand for special public service jobs and increases work 
incentives to those in the $4,000-$15,000 range. (See table 6.) Paradox- 
ically, a simplification of the benefit schedule made before PBJI 
reached the Congress significantly increased the demand for public 
service jobs and total program costs. (See table 7.) 

Finally, while Federal incentives to encourage State supplementa- 
tion greatly complicate the PBJI, the}' effectively hold down Federal 
budgetary costs and forestall a substantial loss of benefits among 
current welfare beneficiaries. An alternative arrangement — raising 
Federal benefits by the amount of the Federal supplementation cost — 
increases the relative position of recipients in States with low current 
benefit levels, grants less fiscal relief to current high benefit States, 
and leaves man}' current recipients in high benefit States worse off 
(unless, of course, States would supplement the Federal benefit level 
in the absence of a financial inducement). (See table 11.) 



UNIVERSITY OF FLORIDA 

II I Mill III II III 

3 1262 09118 6162 

ate the sensitivity of 

BJI. Tin 
which \ . no sense exhaustive, and num ro 

• them are the folio* .. 
I people in ©qua] need recen 

ove the Incomes 
[s Dill. A most approprial y 

r the proposed cash assistance progi ai W 

tnd distributional effects of 1'IUI if tnploy- 

men! rate is abov< m the 1981 unemployment i 

ted by the administratis